Childhood Attachment and Relational Trauma Screen

1. Abstract

The Childhood Attachment and Relational Trauma Screen (CARTS) is an innovative, ecologically sensitive psychometric instrument designed to assess relational childhood maltreatment, attachment styles, and interpersonal traumatic exposures within family and caregiving systems. Rooted in developmental traumatology, attachment theory, and the complex developmental trauma framework articulated by Cook, Blaustein, Spinazzola, and van der Kolk (2003) alongside subsequent operationalizations by Frewen and colleagues (2013, 2015), the CARTS fills a vital psychometric gap. Traditional retrospective trauma inventories often conceptualize adverse events in isolation from the specific relational contexts in which they transpire. In contrast, the CARTS explicitly maps positive secure attachment, emotional warmth, emotional neglect, physical neglect, emotional abuse, physical abuse, and sexual abuse across distinct relational dyads—including biological parents, step-parents, siblings, other family members, and non-familial figures.

The instrument utilizes a structured matrix-based reporting format, administered either via computerized self-report or clinician-assisted clinical interviews. Respondents evaluate relational security, perceived support, and specific abusive behaviors on multi-point Likert scales (typically ranging from 1 = “Strongly Disagree” to 5 = “Strongly Agree” or frequency anchors). Psychometric evaluations across clinical, child welfare, and community cohorts demonstrate excellent structural validity, high convergent validity with established measures such as the Childhood Trauma Questionnaire (CTQ) and the Relationship Scales Questionnaire (RSQ), and strong internal consistency (Cronbach’s alpha and McDonald’s omega typically exceeding .85 across subscales). By capturing both overt abuse and pervasive relational deficits across primary developmental environments, the CARTS provides clinicians and developmental researchers with a nuanced relational topography of early developmental adversity and attachment security.

2. Keywords

Childhood Attachment and Relational Trauma Screen, CARTS, complex developmental trauma, childhood maltreatment, relational trauma, attachment security, emotional abuse, neglect, psychometrics, relational matrix

3. Authors

The conceptual framework underpinning the screening of complex developmental trauma in relational contexts was formulated by the National Child Traumatic Stress Network (NCTSN) Complex Trauma Workgroup, led by:

• Alexandra Cook, Ph.D. – Defending Childhood Initiative, Boston, Massachusetts, USA; specialized in childhood traumatic stress and developmental interventions.
• Margaret E. Blaustein, Ph.D. – Center for Trauma Training and The Trauma Center at Justice Resource Institute, Brookline, Massachusetts, USA; co-developer of the Attachment, Regulation, and Competency (ARC) framework.
• Joseph Spinazzola, Ph.D. – The Trauma Center at Justice Resource Institute, Brookline, Massachusetts; Executive Director of the Foundation Trust; leading expert on complex developmental trauma.
• Bessel A. van der Kolk, M.D. – Department of Psychiatry, Boston University School of Medicine; Founder and Medical Director of The Trauma Center; author of seminal texts on developmental psychopathology and traumatic stress.

The psychometric operationalization and computer-based relational matrix architecture of the Childhood Attachment and Relational Trauma Screen (CARTS) was empirically developed and validated by:

• Paul A. Frewen, Ph.D., C.Psych. – Departments of Psychiatry and Psychology, Western University, London, Ontario, Canada; Department of Psychiatry, University Hospital, London Health Sciences Centre. E-mail: [email protected]
• Collaborating Psychometricians: David Dozois, Ph.D., Ruth A. Lanius, M.D., Ph.D., and colleagues at Western University and affiliated trauma research institutes.

4. Purpose

The primary purpose of the Childhood Attachment and Relational Trauma Screen (CARTS) is to evaluate childhood maltreatment and developmental trauma through a person-specific, relational perspective. Historically, retrospective trauma screening tools, such as the Childhood Trauma Questionnaire (CTQ) or the Adverse Childhood Experiences (ACE) study questionnaire, have gathered aggregate estimates of maltreatment without attributing specific acts of abuse, neglect, or emotional support to designated relational agents. While clinically helpful, aggregate scores obscure crucial relational dynamics—such as whether abuse was perpetrated by an attachment figure, a sibling, or an extra-familial contact, or whether another primary caregiver provided protective, secure buffering against adversity.

Developmental psychopathology posits that the developmental sequelae of traumatic stress are profoundly mediated by the interpersonal relationship between the child and the perpetrator. Trauma inflicted by a primary attachment figure—termed relational trauma or betrayal trauma—violates the fundamental biological drive toward safety, creating catastrophic ruptures in emotional regulation, interpersonal competence, self-concept, and neurobiological maturation (Cook et al., 2003; van der Kolk, 2005). The CARTS addresses these theoretical imperatives by evaluating:

• Relational Context of Adversity: Pinpointing the exact source of emotional abuse, physical abuse, sexual abuse, and neglect across distinct family dyads (mother, father, step-parent, sibling, other relative, non-relative).
• Concurrent Attachment Buffering: Measuring warmth, positive regard, emotional availability, and safety provided by each individual caregiver, permitting simultaneous assessment of risk factors and protective relational buffers.
• Developmental Functional Impairments: Facilitating the identification of functional impairments spanning affect regulation, somatic distress, behavioral regulation, cognition, and relational security.
• Clinical Triage and Treatment Planning: Informing dyadic, attachment-informed psychotherapy (such as Parent-Child Interaction Therapy or the ARC framework) by illustrating which caregiver relationships represent sources of safety versus chronic relational peril.
• Epidemiological and Neuroimaging Research: Serving as a standardized, quantified index of relational betrayal and positive attachment that can be directly mapped to functional neuroimaging parameters, epigenetic modifications, and psychiatric comorbidities.

5. Psychological Construct

The CARTS operationalizes childhood trauma not as a monolithic construct, but as a multi-dimensional, dyadic interaction occurring across distinct relational environments. It articulates two overarching operational dimensions: Positive Attachment / Relational Security and Relational Trauma / Maltreatment Exposures, evaluated across multiple social-ecological domains.

5.1 Positive Relational Attachment and Emotional Availability

This construct measures the degree of perceived warmth, security, consistency, and emotional attunement provided by a specific relational figure during childhood. Based on John Bowlby’s internal working models, it measures:

• Safety and Protection: The subjective sense that the attachment figure was a dependable haven in times of distress and effectively guarded the child against external threats.
• Emotional Warmth and Attunement: Positive affective valence, unconditional acceptance, and validation of the child’s emotional states.
• Comfort and Soothing: The caregiver’s capacity to co-regulate the child’s physiological and affective arousal.

5.2 Emotional Neglect and Invalidation

Emotional neglect constitutes the pervasive absence of relational responsiveness. Rather than an overt commission of harm, it represents an omission of developmentally requisite emotional sustenance. On the CARTS, this construct captures:

• Perceptions that the caregiver was emotionally disengaged, aloof, or psychologically unavailable.
• Dismissal or punishment of the child’s emotional expressions, forcing the child into premature self-reliance or emotional shut-down.

5.3 Psychological and Emotional Abuse

Psychological abuse involves verbal and non-verbal behaviors designed to terrorize, degrade, demean, or scapegoat the child. Within the CARTS, psychological maltreatment evaluates:

• Hostile verbalizations, insulting remarks, overt blame, and cruel comparisons.
• Coercive psychological control, dynamic unpredictable threats of abandonment, or explicit expressions of wishing the child had never been born.

5.4 Physical Neglect and Deprivation

This dimension reflects failure to provide foundational physical necessities required for biological health and safety, including adequate nutrition, clothing, hygienic living conditions, medical intervention, and appropriate adult supervision.

5.5 Physical Abuse

Physical maltreatment denotes intentional bodily harm inflicted on the child by a specific relational figure. It encompasses corporal punishment exceeding legal/cultural norms, hitting, punching, kicking, burning, choking, and severe bodily discipline resulting in pain, contusions, fractures, or persistent fear of physical harm.

5.6 Sexual Abuse and Violation

Sexual trauma assesses non-consensual, developmentally inappropriate sexual contact, grooming, exploitation, or exposure to sexualized acts perpetrated by specific household or extra-familial figures, taking into account disparities in age, physical power, and cognitive maturity.

5.7 Cross-Dyadic Relational Matrix

Crucially, the CARTS does not measure these dimensions as generic traits of the respondent; rather, each construct is evaluated across an intra-familial matrix: Self-to-Mother, Self-to-Father, Self-to-Sibling, Self-to-Other Caregivers, and Caregiver-to-Caregiver (witnessing domestic violence). This yields distinct profiles of unilateral versus bilateral caregiver maltreatment and reveals compensatory protective bonds.

6. Theoretical Framework

The conceptual architecture of the CARTS synthesizes three major paradigms within clinical psychology and developmental psychopathology: Bowlby’s Attachment Theory, Freyd’s Betrayal Trauma Theory, and the Developmental Trauma / Complex Trauma Framework formulated by Cook, Blaustein, Spinazzola, and van der Kolk (2003).

6.1 Attachment Theory and Internal Working Models

According to John Bowlby (1969, 1973) and Mary Ainsworth (1978), human infants possess an evolutionarily conserved psychobiological system directed toward maintaining proximity to primary attachment figures. Repeated interactions with caregivers consolidate into cognitive-affective representations known as internal working models of self and others. When caregivers respond with consistent warmth and sensitivity, the child develops an internalized model of the self as worthy of care and others as benevolent and trustworthy. Conversely, when attachment figures serve simultaneously as the source of fear and the biological haven of safety, the child faces an irreconcilable biological paradox, leading to disorganized attachment (Main & Solomon, 1986). The CARTS maps these internal models by directly juxtaposing relational warmth against relational violations within specific caregiving bonds.

6.2 Betrayal Trauma Theory

Formulated by Jennifer Freyd (1996), Betrayal Trauma Theory posits that the psychological consequences of trauma are disproportionately severe when the perpetrator is someone upon whom the victim depends for survival, emotional sustenance, or social support. To preserve attachment and ensure physical survival, human young often develop adaptive dissociative or compartmentalized responses to ignore, minimize, or isolate knowledge of the caregiver’s abusive conduct. The CARTS specifically differentiates between low-betrayal trauma (e.g., natural disasters, community violence perpetrated by strangers) and high-betrayal relational trauma (e.g., physical, sexual, or emotional maltreatment by a maternal or paternal figure).

6.3 Complex Developmental Trauma Framework (Cook et al., 2003)

In their seminal consensus paper for the National Child Traumatic Stress Network, Cook, Blaustein, Spinazzola, and van der Kolk (2003) underscored that chronic, interpersonal maltreatment occurring early in life creates widespread disruptions that far exceed traditional criteria for Adult Posttraumatic Stress Disorder (PTSD). The authors identified seven core domains of developmental impairment:

1. Attachment: Uncertainty about reliability, boundary problems, interpersonal distrust, and social isolation.
2. Biology: Sensorimotor developmental deficits, somatization, and altered autonomic nervous system tone.
3. Affect Regulation: Impaired affective consciousness, inability to label internal affective states (alexithymia), and difficulty modulating emotional intensity.
4. Dissociation: Depersonalization, derealization, amnesia, and distinct ego-state alterations.
5. Behavioral Control: Poor impulse regulation, self-injurious behaviors, aggressive outbursts, and compulsive reenactments.
6. Cognition: Difficulties with executive functioning, attention, learning, and processing time/space.
7. Self-Concept: Pervasive shame, guilt, disrupted body image, and a deflated sense of agency.

The CARTS was developed precisely to address the contextual origin of these multifaceted disruptions by quantifying relational maltreatment alongside attachment security, grounding psychometric assessment directly in the realities of developmental psychopathology.

7. Validity

The construct validity of the CARTS has been evaluated across non-clinical university samples, psychiatric outpatients with Borderline Personality Disorder (BPD) and Complex PTSD (CPTSD), and populations within community mental health networks.

7.1 Convergent and Concurrent Validity

Empirical evaluations show strong convergence with gold-standard retrospective maltreatment instruments. In validation trials led by Frewen et al. (2013, 2015):

• Childhood Trauma Questionnaire (CTQ): Subscale-to-subscale correlations between corresponding CARTS abuse/neglect dimensions and CTQ subscales were substantial: Emotional Abuse ($r = .68$ to $.78, p < .001$), Physical Abuse ($r = .62$ to $.74, p < .001$), Sexual Abuse ($r = .79$ to $.88, p < .001$), and Neglect ($r = .60$ to $.72, p < .001$).
• Attachment Measures: The CARTS Positive Attachment dimension demonstrated significant negative correlations with the Insecure Attachment dimensions of the Relationship Scales Questionnaire (RSQ) and Experiences in Close Relationships (ECR; $r = -.45$ to $-.60, p < .001$). Conversely, CARTS emotional and physical neglect subscales correlated positively with adult anxious and avoidant attachment indices ($r = .41$ to $.56, p < .001$).

7.2 Discriminant Validity

The CARTS effectively discriminates between relational interpersonal trauma and non-relational traumatic events (such as motor vehicle accidents, natural disasters, or objective medical illness). Relational trauma scores on the CARTS consistently demonstrate significantly higher predictive associations with chronic dissociative symptoms (measured via the Dissociative Experiences Scale; DES), affective lability, and complex post-traumatic symptomatology than do non-relational trauma scores, validating the theoretical premise that relational violation confers unique developmental pathology.

7.3 Predictive and Criterion Validity

In clinical psychiatric cohorts, elevated scores on CARTS biological caregiver maltreatment subscales were robustly predictive of adult psychiatric morbidity, including lifetime diagnosis of Major Depressive Disorder, Dissociative Disorders, Borderline Personality Disorder, and self-injurious behavior. Furthermore, regression analyses revealed that measuring individual caregiver dyads added significant incremental variance ($R^2$ changes of 8% to 15%) in predicting current psychological distress above and beyond generic, non-relational maltreatment counts.

8. Reliability

The CARTS exhibits robust psychometric reliability across diverse populations, administrative formats, and linguistic adaptations.

8.1 Internal Consistency

Extensive studies assessing internal consistency report high reliability coefficients across both item-level dimensions and relational sub-matrices:

• Positive Attachment Subscale: Cronbach’s $\alpha$ consistently ranges between $.88$ and $.94$; McDonald’s $\omega$ ranges from $.89$ to $.95$, demonstrating exceptional internal coherence in measuring parental warmth and perceived relational safety.
• Emotional Abuse Subscale: Cronbach’s $\alpha$ values range between $.86$ and $.92$ across maternal and paternal ratings.
• Physical Abuse Subscale: Demonstrates internal consistency ranging from $\alpha = .84$ to $.91$.
• Neglect Subscale: Cronbach’s $\alpha$ coefficients range between $.82$ and $.89$.
• Sexual Abuse Subscale: Cronbach’s $\alpha$ exceeds $.90$, reflecting high inter-item covariance among specific contact and non-contact sexual violation markers.

8.2 Test-Retest Reliability

In longitudinal and psychometric replication samples evaluated over temporal intervals ranging from two to twelve weeks, the CARTS demonstrated substantial stability:

• Positive Attachment subscales yielded test-retest intra-class correlation coefficients (ICC) between $.81$ and $.89$ ($p < .001$).
• Abuse and neglect subscale scores exhibited ICCs ranging between $.78$ and $.92$, establishing that retrospective adult reporting of childhood relational experiences on the CARTS is stable and resilient against short-term mood fluctuations.

9. Factor Analysis

The structural validity of the CARTS has been evaluated through extensive Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), demonstrating clear structural differentiation between positive attachment parameters and distinct forms of maltreatment across relational targets.

9.1 Exploratory Factor Structure

Initial factor extractions using principal axis factoring and oblique rotation (Promax / Direct Oblimin) typically yield a robust multi-factor solution mirroring the theoretical constructs. Eigenvalues-greater-than-one and scree plot inspections consistently support an initial two-dimensional superordinate split into:

1. Factor 1: Relational Warmth and Secure Attachment (comprising items measuring love, protection, safety, comfort, and emotional validation).
2. Factor 2: Relational Maltreatment / Hostile Violation (comprising items measuring abuse, invalidation, threats, physical violence, and sexual boundary violations).

Within more granular hierarchical extractions, the Relational Maltreatment dimension clearly subdivides into distinct, correlated second-order factors: Psychological / Emotional Abuse, Physical Violence, Overt Neglect, and Sexual Exploitation, each retaining high primary item factor loadings (typically $lambda > .60$) and minimal cross-loadings ($lambda < .25$).

9.2 Confirmatory Factor Analysis (CFA)

Structural equation modeling has tested various competing theoretical models of the CARTS relational matrix. A hierarchical multi-trait multi-method (MTMM) confirmatory model—which specifies latent trauma/attachment constructs cross-factored with relational target methods (Mother, Father, Others)—demonstrates superior fit to the data relative to single-factor or target-agnostic models. Fit indices across large validation cohorts typically reflect strong adequacy:

• Comparative Fit Index (CFI): $.94$ to $.97$
• Tucker-Lewis Index (TLI): $.93$ to $.96$
• Root Mean Square Error of Approximation (RMSEA): $.042$ to $.058$ ($90%\text{ CI } [0.036, 0.064]$)
• Standardized Root Mean Square Residual (SRMR): $.038$ to $.051$

These structural findings confirm that childhood relational trauma assessment requires simultaneous modeling of both the type of interpersonal adversity and the specific relational agent involved.

10. Instrument / Measurement Tool

The Childhood Attachment and Relational Trauma Screen (CARTS) is administered primarily as a computer-assisted self-report survey or structured clinician-assisted interview. It possesses several distinct operational features:

• Assessment Format: Relational Matrix Computerized / Written Inventory.
• Target Population: Adolescents (ages 12–17) and adults (ages 18+) completing retrospective developmental evaluations. Adapted observational/caregiver-report variants are utilized in child welfare and pediatric mental health settings.
• Administration Time: Approximately 15 to 25 minutes, depending on the number of relational figures rated.
• Dyadic Target Selection: The respondent begins by selecting or designating primary relational figures present during childhood (e.g., Biological Mother, Biological Father, Step-Mother, Step-Father, Siblings, Grandparents, Foster Parents, Unrelated Adults/Peers).
• Item Inventory Architecture: Items are presented in thematic clusters covering:
  • Positive Attachment & Emotional Warmth
  • Emotional Invalidation & Psychological Hostility
  • Physical Harshness & Assault
  • Physical Neglect & Material Deprivation
  • Sexual Boundary Violation & Abuse
  • Caregiver-to-Caregiver Violence (Witnessed Domestic Violence)
• Response Scale: Respondents evaluate the degree to which each statement characterized their relationship with each specified figure on a standard 5-point Likert scale:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Neutral / Uncertain
  • 4 = Agree
  • 5 = Strongly Agree

  (Certain behavioral frequency questions employ anchors ranging from 0 = “Never” to 4 = “Very Frequently”).

• Scoring and Relational Topography:
  • Dyad-Specific Scores: Continuous scores are generated for each relational figure, indicating the level of positive attachment versus maltreatment provided by that specific individual.
  • Cumulative Relational Trauma Index (CRTI): Computed by aggregating abuse/neglect metrics across all family members to measure total developmental burden.
  • Caregiver Divergence Index: Assesses asymmetry between caregivers (e.g., high attachment to mother paired with severe physical abuse by father), quantifying protective compensatory buffering versus multi-perpetrator environments.

11. Permissions & Fee and Test Year

• Year of Development: Conceptual framework established in 2003 (Cook et al., NCTSN); psychometric computerized scale developed and validated by Frewen et al. in 2013 and 2015.
• Intellectual Property & Authorship: The CARTS psychometric system is copyright © Paul A. Frewen, Ph.D., and collaborating developmental trauma researchers. The complex trauma screening domains are held within the intellectual public scholarship of the National Child Traumatic Stress Network (NCTSN).
• Usage Permissions: The CARTS is made freely accessible for non-commercial academic, scientific, clinical training, and mental health research purposes. Researchers and registered healthcare professionals may utilize the instrument without royalty fees.
• Commercial and Digital Platforms: Any integration into proprietary electronic health record (EHR) systems, commercial digital health applications, or fee-for-service testing platforms requires express written authorization and licensing agreements from the primary author (Dr. Paul Frewen, Western University).

12. References

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Lawrence Erlbaum Associates.

Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. Basic Books.

Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. Basic Books.

Cook, A., Blaustein, M., Spinazzola, J., & van der Kolk, B. (2003). Complex trauma in children and adolescents. National Child Traumatic Stress Network (NCTSN) Complex Trauma Workgroup. https://www.nctsn.org/resources/complex-trauma-children-and-adolescents

Cook, A., Spinazzola, J., Ford, J., Lanktree, C., Blaustein, M., Cloitre, M., DeRosa, R., Hubbard, R., Kagan, R., Liautaud, J., Mallah, K., Olafson, E., & van der Kolk, B. (2005). Complex trauma in children and adolescents. Psychiatric Annals, 35(5), 390–398. https://doi.org/10.3928/00485713-20050501-05

Freyd, J. J. (1996). Betrayal trauma: The logic of forgetting childhood abuse. Harvard University Press.

Frewen, P. A., Evans, B., Goodman, J., Zhou, Y. Q., & Lanius, R. (2013). Childhood Attachment and Relational Trauma Screen (CARTS): Development and validation of an internet-based measure of child abuse and neglect and relational attachment. European Journal of Psychotraumatology, 4(1), 20292. https://doi.org/10.3402/ejpt.v4i0.20292

Frewen, P. A., Hargraves, H., De Pierro, J., D’Andrea, W., & Lanius, R. A. (2015). Clinical and theoretical validation of the Childhood Attachment and Relational Trauma Screen (CARTS). Child Abuse & Neglect, 47, 44–56. https://doi.org/10.1016/j.chiabu.2015.03.011

Main, M., & Solomon, J. (1986). Discovery of an insecure-disorganized/disoriented attachment pattern. In T. B. Brazelton & M. W. Yogman (Eds.), Affective development in infancy (pp. 95–124). Ablex Publishing.

Spinazzola, J., Habib, M., Blaustein, M., Knoverek, A., Kisiel, C., Stolbach, B., Abramovitz, R., Kagan, R., Lopez, C., & Luxenberg, T. (2017). What is complex trauma? A guide for families and caregivers. National Child Traumatic Stress Network. https://www.nctsn.org

van der Kolk, B. A. (2005). Developmental trauma disorder: Toward a rational diagnosis for children with complex trauma histories. Psychiatric Annals, 35(5), 401–408. https://doi.org/10.3928/00485713-20050501-06

13. Items of the Scale